1GIT-BLAME(1) Git Manual GIT-BLAME(1)
2
6 git-blame - Show what revision and author last modified each line of a
7 file
8
10 git-blame [-c] [-b] [-l] [--root] [-t] [-f] [-n] [-s] [-p] [-w] [--incremental] [-L n,m]
11 [-S <revs-file>] [-M] [-C] [-C] [--since=<date>]
12 [<rev> | --contents <file>] [--] <file>
13
15 Annotates each line in the given file with information from the
16 revision which last modified the line. Optionally, start annotating
17 from the given revision.
18 Also it can limit the range of lines annotated.
20 This report doesn´t tell you anything about lines which have been
22 deleted or replaced; you need to use a tool such as git-diff(1) or the
23 "pickaxe" interface briefly mentioned in the following paragraph.
24 Apart from supporting file annotation, git also supports searching the
26 development history for when a code snippet occurred in a change. This
27 makes it possible to track when a code snippet was added to a file,
28 moved or copied between files, and eventually deleted or replaced. It
29 works by searching for a text string in the diff. A small example:
30 $ git log --pretty=oneline -S´blame_usage´
34 5040f17eba15504bad66b14a645bddd9b015ebb7 blame -S <ancestry-file>
35 ea4c7f9bf69e781dd0cd88d2bccb2bf5cc15c9a7 git-blame: Make the output
36
39 -b
40 Show blank SHA-1 for boundary commits. This can also be controlled
41 via the blame.blankboundary config option.
42 --root
44 Do not treat root commits as boundaries. This can also be
45 controlled via the blame.showroot config option.
46 --show-stats
48 Include additional statistics at the end of blame output.
49 -L <start>,<end>
51 Annotate only the given line range. <start> and <end> can take one
52 of these forms:
53 · number
56 If <start> or <end> is a number, it specifies an absolute line
58 number (lines count from 1).
59 · /regex/
61 This form will use the first line matching the given POSIX
63 regex. If <end> is a regex, it will search starting at the line
64 given by <start>.
65 · +offset or -offset
67 This is only valid for <end> and will specify a number of lines
69 before or after the line given by <start>.
70 -l
72 Show long rev (Default: off).
73 -t
75 Show raw timestamp (Default: off).
76 -S <revs-file>
78 Use revs from revs-file instead of calling git-rev-list(1).
79 -p, --porcelain
81 Show in a format designed for machine consumption.
82 --incremental
84 Show the result incrementally in a format designed for machine
85 consumption.
86 --contents <file>
88 When <rev> is not specified, the command annotates the changes
89 starting backwards from the working tree copy. This flag makes the
90 command pretend as if the working tree copy has the contents of he
91 named file (specify - to make the command read from the standard
92 input).
93 -M|<num>|
95 Detect moving lines in the file as well. When a commit moves a
96 block of lines in a file (e.g. the original file has A and then B,
97 and the commit changes it to B and then A), traditional blame
98 algorithm typically blames the lines that were moved up (i.e. B) to
99 the parent and assigns blame to the lines that were moved down
100 (i.e. A) to the child commit. With this option, both groups of
101 lines are blamed on the parent.
102 <num> is optional but it is the lower bound on the number of
104 alphanumeric characters that git must detect as moving within a
105 file for it to associate those lines with the parent commit.
106 -C|<num>|
108 In addition to -M, detect lines copied from other files that were
109 modified in the same commit. This is useful when you reorganize
110 your program and move code around across files. When this option is
111 given twice, the command looks for copies from all other files in
112 the parent for the commit that creates the file in addition.
113 <num> is optional but it is the lower bound on the number of
115 alphanumeric characters that git must detect as moving between
116 files for it to associate those lines with the parent commit.
117 -h, --help
119 Show help message.
120 -c
122 Use the same output mode as git-annotate(1) (Default: off).
123 --score-debug
125 Include debugging information related to the movement of lines
126 between files (see -C) and lines moved within a file (see -M). The
127 first number listed is the score. This is the number of
128 alphanumeric characters detected to be moved between or within
129 files. This must be above a certain threshold for git-blame to
130 consider those lines of code to have been moved.
131 -f, --show-name
133 Show filename in the original commit. By default filename is shown
134 if there is any line that came from a file with different name, due
135 to rename detection.
136 -n, --show-number
138 Show line number in the original commit (Default: off).
139 -s
141 Suppress author name and timestamp from the output.
142 -w
144 Ignore whitespace when comparing parent´s version and child´s to
145 find where the lines came from.
146
148 In this format, each line is output after a header; the header at the
149 minimum has the first line which has:
150 · 40-byte SHA-1 of the commit the line is attributed to;
153 · the line number of the line in the original file;
155 · the line number of the line in the final file;
157 · on a line that starts a group of line from a different commit than
159 the previous one, the number of lines in this group. On subsequent
160 lines this field is absent.
161 This header line is followed by the following information at least once
162 for each commit:
163 · author name ("author"), email ("author-mail"), time
166 ("author-time"), and timezone ("author-tz"); similarly for
167 committer.
168 · filename in the commit the line is attributed to.
170 · the first line of the commit log message ("summary").
172 The contents of the actual line is output after the above header,
173 prefixed by a TAB. This is to allow adding more header elements later.
174
176 Unlike git-blame and git-annotate in older git, the extent of
177 annotation can be limited to both line ranges and revision ranges. When
178 you are interested in finding the origin for ll. 40-60 for file foo,
179 you can use -L option like these (they mean the same thing — both ask
180 for 21 lines starting at line 40):
181 git blame -L 40,60 foo
184 git blame -L 40,+21 foo
185 Also you can use regular expression to specify the line range.
186 git blame -L ´/^sub hello {/,/^}$/´ foo
189 would limit the annotation to the body of hello subroutine.
190 When you are not interested in changes older than the version v2.6.18,
192 or changes older than 3 weeks, you can use revision range specifiers
193 similar to git-rev-list:
194 git blame v2.6.18.. -- foo
197 git blame --since=3.weeks -- foo
198 When revision range specifiers are used to limit the annotation, lines
199 that have not changed since the range boundary (either the commit
200 v2.6.18 or the most recent commit that is more than 3 weeks old in the
201 above example) are blamed for that range boundary commit.
202 A particularly useful way is to see if an added file have lines created
204 by copy-and-paste from existing files. Sometimes this indicates that
205 the developer was being sloppy and did not refactor the code properly.
206 You can first find the commit that introduced the file with:
207 git log --diff-filter=A --pretty=short -- foo
210 and then annotate the change between the commit and its parents, using
211 commit^! notation:
212 git blame -C -C -f $commit^! -- foo
215
217 When called with --incremental option, the command outputs the result
218 as it is built. The output generally will talk about lines touched by
219 more recent commits first (i.e. the lines will be annotated out of
220 order) and is meant to be used by interactive viewers.
221 The output format is similar to the Porcelain format, but it does not
223 contain the actual lines from the file that is being annotated.
224 1. Each blame entry always starts with a line of:
227 <40-byte hex sha1> <sourceline> <resultline> <num_lines>
230 Line numbers count from 1.
231 2. The first time that commit shows up in the stream, it has various
233 other information about it printed out with a one-word tag at the
234 beginning of each line about that "extended commit info" (author,
235 email, committer, dates, summary etc).
236 3. Unlike Porcelain format, the filename information is always given
238 and terminates the entry:
239 "filename" <whitespace-quoted-filename-goes-here>
242 and thus it´s really quite easy to parse for some line- and
243 word-oriented parser (which should be quite natural for most
244 scripting languages).
245 Note
247 For people who do parsing: to make it more robust, just ignore any
248 lines in between the first and last one ("<sha1>" and "filename"
249 lines) where you don´t recognize the tag-words (or care about that
250 particular one) at the beginning of the "extended information"
251 lines. That way, if there is ever added information (like the
252 commit encoding or extended commit commentary), a blame viewer
253 won´t ever care.
254
257 git-annotate(1)
258
260 Written by Junio C Hamano <junkio@cox.net>
261
263 Part of the git(7) suite
264Git 1.5.3.3 10/09/2007 GIT-BLAME(1)